LLM Reference

Qwen3.6 Max Preview

Released
2026-04-20
Last refreshed
2026-06-29
Status
Researched 120d ago
Open sourceCommercial use: permittedMultimodalCodingRAGAgentsLong contextVisionClassificationJSON / Tool use

Qwen3.6 Max Preview is worth evaluating for coding, rag, and agents when its provider route and context window match the workload.

Use it for

  • Teams evaluating coding, rag, and agents
  • Workloads that can use a 256k context window
  • Buyers comparing 3 tracked provider routes

Do not use it for

  • Workloads where another current model has stronger sourced task evidence
Specifications
Family
Qwen3.6
Released
2026-04-20
Context
256k
Architecture
Mixture of Experts
Specialization
general
Openness
Open source
License
Apache 2.0OSI-approvedCommercial use: permitted
Weights
Unknown
Code
Unknown
Training
Pretrained
Created by
AlibabaAlibaba

AI research institute of Alibaba Group.

Hangzhou, Zhejiang, China
Founded 2017
Website
Pricing
Output / 1M
$6.24
Input / 1M
$1.04

Cheapest of 3 routes · Alibaba Cloud PAI-EAS

Providers(3)
OpenRouterAlibaba Cloud PAI-EASVercel AI Gateway
View 3 provider routes
Links
Website

About

Qwen3.6-Max-Preview is a proprietary frontier model from Alibaba Cloud built on a sparse MoE architecture, available for preview as part of the Qwen3.6 series.

Top use-case fit: coding, agents, and build tasks

Coding

Q/$ D

2 relevant benchmarks in the decision map.

RAG

Included by capability and metadata signals in the decision map.

Agents

Q/$ C

1 relevant benchmark in the decision map.

Provider price ladder

Compare all 3

Compare API pricing across 3 providers for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MCacheRoute
Alibaba Cloud PAI-EAS$1.04$6.24-
Serverless
OpenRouter$1.04$6.24-
Serverless
Vercel AI Gateway$1.30$7.80read $0.260
Serverless

Capabilities

VisionMultimodalReasoningJSON / Tool useStructured Outputs

Benchmark peer barsfor Coding

SWE-bench VerifiedRank 22 of 81
Claude Fable 5
96.0
Claude Opus 5
96.0
Claude Mythos Preview
93.9
Claude Opus 4.8
88.6
Qwen3.6 Max Previewcurrent
78.8
LiveCodeBenchRank 14 of 56
DeepSeek V4 Pro
93.5
Gemini 3.1 Pro Preview
91.7
DeepSeek V4 Flash
91.6
Qwen3.7-Max
91.6
Qwen3.6 Max Previewcurrent
87.1

Benchmark scores(6)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionEvaluationSource
Google-Proof Q&A86.0diamondObserved 2026-05-12—Source
Massive Multi-discipline Multimodal Understanding82.0MMMUObserved 2026-05-12—Source
Humanity's Last Exam28.8HLE Thinking Level Max (no tools), from DataLearner (accuracy)Observed 2026-06-07—Source
LiveCodeBench87.1Ranked 10th of 120 on LiveCodeBench from DataLearner (pass@1)Observed 2026-06-07—Source
MMLU PRO88.5MMLU-Pro (accuracy)Observed 2026-06-07—Source
SWE-bench Verified78.8SWE-bench Verified (resolved)Observed 2026-06-07—Source

Migration checks

No linked migration route is available for this model yet.

Compare Qwen3.6 Max Preview with other models

Show all 12 popular comparisonssorted by 7-day search impressions
Qwen3.6 Max Preview vs Grok-311Qwen3.6 Max Preview vs DeepSeek V311Qwen3.6 Max Preview vs Gemini 2.5 Flash Live API10Qwen3.6 Max Preview vs Llama 3.1 405B Instruct10Qwen3.6 Max Preview vs GPT-5.4-Cyber10Qwen3.6 Max Preview vs Claude Opus 4.77Qwen3.6 Max Preview vs Llama 2 13B Chat7Qwen3.6 Max Preview vs Together AI - Llama 3 8B Lite6Qwen3.6 Max Preview vs GPT-5.5-Cyber3Qwen3.6 Max Preview vs Grok 43Qwen3.6 Max Preview vs DeepSeek R1 Lite2Qwen3.6 Max Preview vs Phi-4 Reasoning Vision 15B2